ABSTRACT Despite identification of modifiable behavioral targets for childhood obesity prevention, prevalence of obesity remains historically high in the United States and the most severe forms are increasing among young children. Recent epidemiologic evidence shows that racial/ethnic and socioeconomic disparities in obesity prevalence are widening, and already are apparent by age 24 months. Thus, childhood obesity prevention efforts must address factors beyond individual-level behavioral risk factors and should start in the first months of life. Social determinants of health (SDoH) are increasingly recognized as playing key upstream roles in etiologies of obesity, particularly in disproportionately burdened populations. Therefore, SDoH may be prime targets for interventions to prevent childhood obesity among racial/ethnic minority and low-income populations. Recently, innovative health care models have incentivized integration of SDoH screening and social service referrals into electronic health records (EHRs). While technologic advances allow implementation of individual and neighborhood measures of SDoH into EHRs, a critical gap in understanding which factors most strongly predict obesity ? and thus should be prioritized as clinical measures and intervention targets ? precludes implementation of these technologic advances for obesity prevention interventions. Furthermore, few studies have examined relationships of SDoH measures with growth parameters from birth to age 24 months, a critical period of plasticity and development in which stressors can lead to long-term health consequences. The overall goal of this study is to identify SDoH measures at multiple levels (individual, family, community, and environmental) that show promise for adoption into health IT systems to prevent childhood obesity. We also will examine effects of social services referrals and utilization on associations between SDoH and infant weight outcomes. To achieve these aims, we will leverage an existing EHR-based SDoH screening and bi- directional community-based social services referral system. Patients presenting for routine well child care at a multi-site academic practice that serves a large number of racial/ethnic minority and low-income families, will form the basis of a longitudinal cohort of 1300 infants from birth to age 24 months. Additionally, we will perform in-depth interviews to explore patient, provider, and community stakeholder perceptions of EHR-based SDoH screening and social services referrals. The results of this study could strengthen our understanding of the role of specific multi-level determinants of childhood obesity. It also will provide new information about the effects of clinical-community resource linkages on SDoH and infant growth trajectories. This research will fill critical knowledge gaps to accelerate integration of SDoH measures that most strongly predict unhealthy infant weight gain into health information technology systems. The results of the proposed research will form the basis of future comparative effectiveness research to prevent childhood obesity in disproportionately burdened populations. 1